Background
Experiencing depression as a child or adolescent can have a significant impact on the social, occupational, physical and emotional functioning of the young person [
1,
2]. This is of concern given that by the age of 18 years one in every five people will have experienced depression [
3]. It is crucial to provide optimal treatment for this age group in order to minimise the negative impact of depression on their lives [
4]. Despite this need, a large proportion of those young people experiencing depression will not access treatment [
5,
6], and for those who do, many will experience significant delays before accessing services [
7,
8]. There are a number of factors that may lead to a delay in treatment, including stigma [
9], negative attitudes to, and experiences of, treatment [
10,
11], and availability of services [
12]. Once in treatment, engagement remains a considerable challenge [
13].
In order to maximize the chances of successful engagement, a collaborative approach to treatment decision making has been proposed for mental health treatment in general [
14] and for young people experiencing depression specifically [
15]. Clinical guidelines for the treatment of depression in children and adolescents advocate for the inclusion of young people in learning about and making decisions regarding their treatment (e.g. [
16]). Active involvement in treatment decision making may lead to increased satisfaction with services and therefore improve engagement and clinical outcomes [
17,
18].
Involvement in treatment decision making has been discussed in several ways. At a broad level, involvement can be seen to be passive (e.g. paternalistic model of care provided by a clinician, whereby the clinician shares little information with the client, and the clinician deliberates and makes the final decision about treatment); shared (e.g. the client and clinician share information with each other, both deliberate about this information and choose a treatment option together); or autonomous (e.g. the client may obtain information from the clinician but then deliberates on this information and makes a decision themselves) [
19].
It is important to consider what constitutes involvement. In the above model, the conceptualization of involvement focuses on information sharing and making actual decisions. Entwistle and Watt (2006) [
20] suggest a broadening of what constitutes involvement to include both 1) participation in activities related to decision making (with or without other people) and 2) the way that people think and feel about such activities and people. They describe three existing domains of involvement (communication between clients and clinicians about decision making; efforts made by the client in relation to decision making; and efforts made by the clinician in relation to decision making), and propose that two further areas should be considered in order to create a more meaningful understanding of client involvement: 1) the way in which clients view, and feel about, these first three domains as well as their role in, and contribution to, the decision-making process; and 2) consideration of the feelings of both the client and clinician about their relationship with each other.
It is also important to consider the ways in which clients may wish to be involved in the decision making process. Deber and colleagues (2007) [
21] conceptualise two categories related to making decisions: 1) problem solving, which refers to a scenario where there is only one possible correct answer and is therefore not able to be influenced by client preferences; and 2) decision making, which may be based on problem solving, that can be influenced by client preferences and involves some consideration of the potential pros and cons of different scenarios (e.g. treatment options). For each of these categories, it is proposed that clients can prefer to 'keep' the responsibility themselves, 'share' responsibility with a clinician, or 'hand over' responsibility to a clinician. Combinations of these preferences then fall under categories of preference for involvement in decision making (e.g. passive, shared, autonomous). Recently, a 'shared' model of client involvement in decision making, called shared decision making, has received increased interest in both research and clinical settings.
Shared decision making (SDM) facilitated by the use of decision aids (DAs) has been well tested in areas such as breast cancer treatment decision making [e.g. [
22]] and choosing whether or not to undertake screening for prostate cancer [e.g. [
23]]. The most common model of SDM is that described by Charles and colleagues [
24], which involves three major steps: 1) two parties (e.g. doctor and patient) provide each other with relevant information; 2) these same two parties deliberate on the decision by discussing the treatment options and preference for each option; and 3) a treatment option is selected that is consistent with patient preferences and values. This model was updated in 1999 [
19] to further detail these stages of decision making (e.g. that information exchange comprises flow, direction, type and amount of information), to acknowledge that approaches taken to decision making are not necessarily static and may indeed change over time, to describe sub-categories of decision making approaches that are hybrid versions of the broader paternalistic, shared and autonomous categories, and to make the SDM model more relevant to clinical, research and educational purposes (e.g. helping clinicians to understand the different variations of involvement). Edwards and Elwyn (2006) [
25] have highlighted the importance of focusing on the process of decision making (e.g. presentation of treatment options, information sharing and consideration of preferences for who should make the decision) rather than on who makes the actual decision.
DAs are evidence-based tools that provide information about the potential risks and benefits of treatment options (or other health related choices such as screening tests) and are designed to elicit patient preferences in relation to these options. In doing so, they are also designed to facilitate the exchange of information and increase the amount of information shared between the healthcare provider and consumer. The establishment of the International Patient Decision Aid Standards (IPDAS), means that the quality of DAs can now be assessed [
26].
Despite enthusiasm for the application of SDM for mental disorders [e.g. [
27,
28]] a recent systematic review concluded that only two randomized controlled trials (RCTs) had been conducted investigating SDM for mental disorders [
29]. The first study was a cluster RCT that tested an intervention that involved SDM training for physicians and a DA for adult outpatients diagnosed with depressive disorders [
30]. The intervention significantly increased patient involvement and satisfaction without increasing the duration of consultations; however, the SDM intervention had no impact on level of depression severity. The second study was a single site RCT that investigated SDM with the use of a DA for inpatients diagnosed with schizophrenia [
31,
32]. The intervention demonstrated that SDM was feasible for this population, and significantly increased patients' knowledge about schizophrenia, uptake of psychoeducation, and feelings of involvement in consultations, again, without increasing consultation time. However, as with the first study, clinical symptom severity did not improve.
Despite there only being two RCTs testing interventions specific to SDM and the use of DAs, there have been a number of efforts in recent years that have taken a person-centered approach to interventions for depression and other mental disorders. Most collaborative care models (CCMs) in the US, for example, have incorporated patient-centered decision making processes and have been demonstrated to improve clinical outcomes such as adherence to medication, depression severity, quality of life and patient satisfaction [
33]. CCMs have largely been tested in adults; however, a small number of studies have been reported in young people diagnosed with depressive disorders [
34‐
37]. Of the three studies located, two involved client choice as part of the intervention, and one did not, as described below.
A small pilot study was conducted testing a CCM based on an intervention designed for older adults and adapted for young people 12-18 years being seen in primary care [
34]. The 6-month intervention included client choice of treatment with input from caregivers, and was found to be acceptable to young people, their caregivers and physicians, and depression scores improved for the majority of participants. Two RCTs have been conducted; the first randomized participants to either 12 months of treatment as usual (TAU; predominantly prescription of SSRI medication alone) or TAU plus brief cognitive behavioural therapy (CBT; five to nine sessions), ongoing consultation with a therapist and primary care provider, and follow up contact via phone for 12 months [
35]. Participants were not offered a choice in terms of medication or psychological therapy. A weak effect was found for the CCM intervention, which may have been influenced by the TAU condition being relatively effective, the small sample sizes and the low adherence to selective serotonin reuptake inhibitor (SSRI) medication in the CCM intervention group. The second RCT, the Youth Partners in Care study [
36,
37] designed an intervention aimed to provide young people aged 13-21 years who were being seen in primary care settings with improved access to evidence-based treatments and compared this to usual care. As part of the CCM intervention, participants were informed about, and involved in, making decisions about treatment options. Similar to results of studies with adults, the 6-month intervention significantly improved depression severity, quality of life and patient satisfaction. While there is an increased cost associated with CCMs, they may in fact be a prudent investment given their effectiveness in improving clinical outcomes and the financial costs and losses seen with untreated depression. The results from these studies offer insight into the effects of CCMs, yet it is difficult to tease apart the contribution of the patient-centered elements. Therefore, the effects seen with CCMs compared to SDM only interventions may be somewhat different.
Both SDM approaches and CCM approaches that afford participants treatment choice, assume that clients have a desire to be involved in making decisions about their own treatment and care. A narrative review investigating factors related to patient preference for involvement in decision making concluded that this was influenced by demographic variables, experiences of illness and medical care, diagnosis/health status, type of decision, patient literacy about their condition, attitude towards involvement and relationships with health providers [
38]. Studies investigating preference for involvement in adults diagnosed with mental disorders have found a consistent and strong preference for involvement [
39,
40], and that it is feasible to do so [e.g. [
41]]. Despite a desire for involvement, studies measuring levels of SDM in consultations relating to adults with mental disorders have consistently found low levels of involvement [
40‐
45] and the authors were unable to find any studies that have measured SDM behaviours within consultations with young people diagnosed with mental disorders, nor preference for involvement in young people with mental health disorders. The context for treatment decision making for depression is likely to be different than treatment decision making in other health areas [
46], and for young people even more so. Given this, there is a need to consider factors related specifically to this population.
Qualitative methods have been used to investigate specific aspects of depression care for different populations. Adult participants enrolled in a CCM intervention study were interviewed about their experience of collaborative care, and these data were used to amend and improve the model for future research projects [
47]. Focus groups have been used to investigate the attitudes and preferences of different ethnic groups in relation to treatment for depression [
48], as well as stigmatizing beliefs about depression and help seeking for depression [
49]. These studies have highlighted the importance of stigma on help seeking behaviours and adherence to treatment. Experiences and beliefs about treatment for depression have been explored in interviews with adults [
46] and, of specific relevance to this study, young people aged 14-19 years who were involved in either interviews or a focus group [
50]. Both the adult and adolescent samples wanted more information about depression and the available treatment options, as well as support from their clinicians to make treatment decisions.
Given the paucity of data in the area, particularly for clients seen in tertiary mental health settings, we felt that a descriptive account of young peoples' experiences and beliefs about treatment decision making for depression would be a useful starting point from which further work into SDM and DAs for this population could build upon. We also felt that obtaining accounts from caregivers was imperative given the ages/developmental stages and the likely involvement of caregivers in the decision making processes. The aim of the current study therefore was to investigate the experiences and beliefs of young people who had been diagnosed with depressive disorders and their caregivers. Specifically, of interest was the degree to which young people and their caregivers were involved in treatment decision making, and how involved they wished to be. First and foremost, this study sought to obtain rich descriptive data on the above topics in and of themselves; however, a secondary purpose was to investigate whether or not treatment decision making could be improved at each service and to use the data to inform the development of a decision support tool.
Discussion
The most striking finding from these data is the variability in experiences of and desire for involvement in treatment decision making, both within and across clients, clinicians and services. Yet involvement, at some level, in the decision-making process was important to all clients for a broad range of reasons. This complements results from research into preferences for involvement in adults with mental disorders [
36‐
38], and it would be of benefit to further investigate the preferences for involvement that young people with depressive disorders seen at a larger range of services have, to further increase our knowledge about the generalisability of this finding. Particularly given that young people [e.g. [
55]] and adults diagnosed with depression [e.g. [
56]] have demonstrated increased preferences for involvement. Despite showing a strong desire to be involved in the decision-making process, all clients wanted at least some involvement from their clinicians, which supports a collaborative model rather than an autonomous model [
19]. It also demonstrates that client preferences for involvement do not always fit the three main models of involvement (e.g. paternalistic, shared and autonomous). Therefore, a more flexible understanding of involvement that incorporates more complex combinations of preferences is necessary [
20,
25]. The desire for involvement of caregivers varied across clients. Clients' accounts of what constituted true involvement for them focused on factors related to key aspects of the client-clinician relationship, such as engagement and adherence, as well as patient centered goals such as autonomy and empowerment. This supports the notion that conceptualizations of involvement should acknowledge and consider the views and feelings of clients (and caregivers) about their relationships with clinicians [
21,
57]. In line with recent calls promoting SDM for mental disorders [
14,
15], these data support a focus on involvement in decision-making processes for young people with depressive disorders.
The provision of information was also variable across clinicians and services, yet most clients and caregivers voiced a desire for more information. This is in line with previous research investigating the experiences and preferences of adults and adolescents receiving treatment for depression [
46,
50]. For clients it was important that this information accounted for their values and preferences. Some clients felt unable to ask for information, even though they felt that they didn't have sufficient information to be involved in the decision-making process or understand why a treatment was being offered and/or feel satisfied with the decision-making process. Significant barriers were discussed by clients, both in terms of access to services at all and also in terms of being involved in the decision-making process once gaining access to a service.
There are several reasons why informative resources that promote the inclusion of young people in decision-making processes are difficult to produce and may not be available. Reasons may include: a lack of evidence to base information on and the need to update resources according to the latest evidence; challenges with dissemination (e.g. translating evidence into readily accessible and understandable resources); and barriers to implementation (e.g. enlisting the support of organizations and clinicians). Information resources need time for ongoing development and, therefore, ongoing financial commitments from services. While potentially costly, they would provide a systematic way to ensure the opportunity of involvement of each client. Given the significant problems with help seeking in this population [
7,
8], there is an onus on service providers to employ tactics that maximize engagement and adherence to the chosen treatment option. Past negative treatment experiences have been highlighted elsewhere [
12] as a key factor related to accepting a diagnosis of a depressive disorder and, therefore, impacting on help-seeking behaviour, which was echoed in the findings of this study. The need to choose treatments that are preference based and clinically effective in collaboration with the young person upon initial engagement in a service is twofold; with the hope that the treatment will work first time round, and if this is not achieved, that the young person will be willing to remain engaged and pursue further treatment options.
Caregivers found barriers in gaining access to services for their children, but also barriers to being involved in their care once accepted into services. The issue of the age of their children and associated confidentiality policies was the biggest concern reported by caregivers. Whilst such policies are unlikely to change, the use of decision support tools may be one way in which to either involve caregivers in the decision-making process (if so desired by the client) or communicate to caregivers the rationale for the decision made so that they can at least understand what is happening and why.
The experiences of the clients highlighted gaps in the decision-making process, and clients offered ways in which to improve such processes. Information that was interactive and meaningful was a priority for clients, as was feeling as if they were respected and taken seriously by their clinicians. This priority, in combination with clients' desires to have their personal characteristics, values and preferences considered, clearly supports the use of decision support tools and shared decision making. Given that preference for involvement is likely to change over time, having tools available to use on a repeated basis as decisions are revisited seems warranted. Understanding treatment options both for themselves and also to explain to caregivers if appropriate, was important for clients in order to navigate the complex process of seeking help and engaging with services.
There are several limitations to the current study. Although we undertook the study in order to obtain a rich description of experiences and beliefs based on a purposive sample, the small sample size minimizes the generalisability of these findings. While the participants were recruited from only two services, they had all experienced treatment decision-making at other services and therefore data were obtained for experiences at general practice, enhanced general practice, the public mental health system and private practitioners. We acknowledge, however, that there are likely to be characteristics unique to this sample that may not be present in the broader population. Given that a secondary aim was to look for ways in which decision making could be improved at each service, we wanted to recruit current clients of the services. This meant that recruitment was difficult, as the clinical needs of the young people needed to be prioritized. Efforts were made, however, to review the interview data as they were being collected to ensure that different types of experiences and beliefs were being addressed. Most young people and caregivers were interviewed as they were being discharged from the service, which meant that they could reflect on their time at the service as well as experiences at services prior to attending their current service. Another limitation is that participants were asked to recall events that they had experienced over several years. In line with our approach, however, we were interested in participants' accounts of their experiences rather than what actually happened. We feel that concepts such as involvement can be very subjective and different parties (e.g. doctor and patient) may describe an encounter such as treatment decision making in very different ways. How young people and their caregivers make sense of such encounters can help us to understand and improve aspects such as treatment decision making.
Conclusions
This study is the first to consider the experiences and beliefs of young people and their caregivers about treatment decision making for depressive disorders. Clinical guidelines advocate for the inclusion of young people in such decision-making processes and the current study supports this.
The difficulty that clients reported getting accepted into services demonstrates that there is an onus on services to maximize efforts to engage clients once accepted. Given that clients reported a direct relationship between involvement and outcomes such as engagement, adherence and satisfaction with services, the importance of at least offering clients involvement in the decision-making process was highlighted. This is particularly true for clinicians or services that either precluded involvement or from which clients readily disengaged.
The factors that influence desire for involvement will not always be evident to clinicians and therefore involvement should be negotiated explicitly (rather than assuming the level of involvement that the client desires and/or can cope with) and repeatedly (because desire to be involved is likely to change over time). Caregiver involvement should be negotiated explicitly and on an individual basis. Caregivers should be supported with the necessary information about mental disorders and treatment options, particularly when they are responsible for key tasks outside of the clinical sessions (such as filling prescriptions and monitoring risk levels).
This study fills a gap in the knowledge about the context in which young people diagnosed with depressive disorders find themselves making treatment decisions, and provides the basis on which to build a body of work looking at the needs of such young people. Further advancement of this area, including the development of quality decision support tools to facilitate shared decision making, will open up the possibility of improved decision-making experiences for young people, which has the potential to improve key clinical outcomes for this population.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
MS conceived the project, conducted the interviews, analysed the data and drafted the manuscript under the supervision of SH and AJ. SH and AJ were also involved in subsequent redrafts of the manuscript. All authors read and approved the final manuscript.